The purpose of this application remains the study of the macromolecular pathology of age-related neuronal attrition in the human brain. We seek to understand the origin and pathogenesis of the intraneuronal paired helical filaments (PHF) and the extraneuronal amyloid fibers that form in the brain during normal aging and in Alzheimer's disease (AD). This laboratory developed techniques for isolating PHF from human brain and discovered unusual and unexpected molecular properties of these fibers. The observation that PHF are highly insoluble, protease resistant, high M.W. polymers has led directly to: (1) a method for preparing enriched PHF fractions in yields suitable for further biochemical study; (2) the hypothesis that PHF result from abnormal crosslinking of neuronal proteins by nondisulfide covalent bonds; (3) the production of PHF-specific polyclonal antibodies that show no reaction with normal brain proteins; and (4) the production of a library of monoclonal antibodies to PHF. On the basis of this progress, we propose a series of molecular studies aimed at the full characterization of the protein composition of PHF and senile plaque core amyloid. Highly enriched fractions of neurofibrillary tangles (NFT) or amyloid cores prepared by our published method (1) will be labeled with our PHF or amyloid antibodies and purified by fluorescence-activated cell sorting. EM analyses of preliminary FACS trials clearly demonstrate the feasibility of this new method. Using purified NFT or amyloid cores, several protein chemical strategies will be carried out: (a) amino acid analyses; (b) N-terminal analyses; (c) modification of noncovalent bonds by chaotropic solvents (e.g., GuSCN) followed by sequential enzymatic hydrolyses; (d) CNBr cleavage; (e) partial acid hydrolysis; (f) HPLC separation of peptides derived from (c)-(e), followed by automated sequencing and computer comparison to known proteins (including other amyloids, neurofilaments and viral proteins); (g) identification of any acid-labile or -stable crosslinks; and (h) analysis of nonprotein constituents. Detailed comparisons of PHF and amyloid purified from sporadic AD, familial AD and Down's syndrome/AD will be undertaken. PHF-specific antibodies will also be used to quantitate (in an RIA) the degree of SDS insolubility of PHF in various cases and to search for related soluble proteins in brain, CSF and serum. These studies should yield new information about the origin of PHF and plaque amyloid and thus provide clues to more fundamental molecular changes that precede their formation.